whereismyai
Argmax OSS
Argmax OSS is a suite of on-device inference frameworks providing speech-to-text, speaker diarization, and text-to-speech capabilities for Apple silicon.
Argmax OSS provides a collection of Swift-based inference frameworks designed for local execution on Apple silicon. The suite includes WhisperKit for OpenAI Whisper speech-to-text, SpeakerKit for Pyannote speaker diarization, and TTSKit for Qwen-TTS text-to-speech generation. Developers can integrate these tools into macOS and iOS applications to handle audio processing tasks entirely on-device, eliminating the need for external servers. The frameworks support various model sizes and allow for custom model generation via HuggingFace compatibility. By utilizing Core ML, the SDK ensures efficient audio transcription, diarization, and speech synthesis within the Apple ecosystem.
Argmax OSS at a glance
Key features
- Speech-to-text transcription with Whisper models
- Speaker diarization using Pyannote
- Text-to-speech generation with Qwen-TTS
- Memory-efficient incremental loading for large audio files
- Local server implementation with OpenAI API compatibility
- Real-time audio streaming and playback
- Automatic language detection
Use it for
- Transcribing long-form audio files locally
- Adding text-to-speech voices to iOS and macOS apps
- Identifying speakers in audio recordings
- Building local voice assistants with streaming responses
Alternatives to Argmax OSS
- MLX Audio — It enables fast audio model inference directly on Apple M-series chips.
- FluidAudio — Runs local audio AI models on Apple Neural Engine.
- AssemblyAI — Industry-leading speech models deliver high accuracy and low latency for voice applications.
- WhisperX — It enables 70x faster, batched transcription with high-accuracy alignment.
- sherpa-onnx — Runs complex audio AI tasks offline on local devices and embedded hardware.
- Free ASR — Zero-dependency real-time streaming speech-to-text with no account login required.
- Speech-to-Speech — Fully modular, swappable components with local or hosted execution options.
- Rapid-MLX — Run 250 MLX-native models locally on Apple Silicon with high inference speeds.
- audio.cpp — Runs modern audio models locally with high-performance, native C++ execution.
- Strata Mlx — Runs 125B MoE models on Apple Silicon by streaming experts directly from disk.
- whisper.cpp — It provides high-performance, offline speech recognition on diverse hardware.
- kokoro-onnx — Runs local TTS inference with efficient ONNX runtime performance.
All Argmax OSS alternatives